ANALYTICAL SCIENCES MAY 2004, VOL The Japan Society for Analytical Chemistry

Transcription

1 ANALYTICAL SCIENCES MAY 2004, VOL The Japan Society for Analytical Chemistry 831 Determination of Anion-Exchange Resin Performance Based on Facile Chloride-Ion Monitoring by FIA-Spectrophotometry with Applications to Water Treatment Operation Yong-Sheng LI* and Yi-Ling DONG** *School of Chemical Engineering of Sichuan University, No. 24 Southern Section 1, 1st Ringroad, Chengdu City, Sichuan Province , China **Peninsula Water Treatment Ltd., No. 381 Central Plaza, Huaihai Zhong-road, Shanghai 20020, China Based on a flow-injection spectrophotometry, an automatic analytical method for determination of ppb-level chloride-ion has been established. By use of this method, a novel FIA method for the determination of SBAER performance has also been developed. In this paper, the effects of concentration, dosage, and flow rate of the regenerant on BEC of SBAER were first investigated dynamically by the FIA method. In addition, the flow rate of the sample water and the temperature of the ion exchange resin were also examined. The optimum conditions were obtained: the volume of the regenerant (sodium hydroxide) was 50 ml (0.15 g resin), and its concentration was 3% (w/v); the volumetric flow rates of the regenerant and the sample water were 0.5 ml/min (4.3 m/h) and 1.5 ml/min (13 m/h), respectively. The exchanging temperature was 25 ± 5 C. The method is characterized by the use of a micro resin-column, shorter testing cycle, easy operation, and high reproducibility. The proposed method is approximately 30 times more efficient than the manual method, and it can be used for the exchange performance comparison of various SBAER. (Received March 31, 2003; Accepted November 19, 2003) Introduction With the development of the ion-exchange technique, ionexchange resins have become more and more widely used; in more than 80% of these case, resins are used in the treatment of water. Ion-exchange technique is also an indispensable way to obtain different grades of purified water. The quality of water is important to ensure the safe and economical operation of large capacity power generating sets at power plants. At present, production sections needing purified water request that the resins used not only have high mechanical strength and ionexchange rate, but also have large exchange capacity. So analysts of factories must conduct the performance comparison test for different kinds of ion-exchange resins before deciding which kind of resin to choose for use. The breakthrough exchange capacity (BEC), equilibrium exchange capacity, and exchange rate of ion-exchange resins are the main performance indices in the industry of water treatment. So far, some researchers have conducted studies on the determination of resin exchange performance, 1 3 but many users have been adopting the manual methods 4 7 to determine the exchange-resin performance. However, the method s shortcomings are difficult operation, long testing time and large consumption of resins and reagent. So it is necessary to develop a simple, quick method for determining the performance of ionexchange resin. For the above reasons, an automatic method for rapidly To whom correspondence should be addressed. lysgxf@mail.jl.cn determining the exchange performance of strongly basic anionexchange resin (SBAER) has been developed in this study based on flow-injection spectrophotometry It is a pathfinder in the aspects of theory and of rapid determination of ionexchange resin performance, and thus has important significance. Experimental Theory Figure 1(a) shows the schematic drawing of an ion exchange resin column. When the chloride solution passes through the column, the SBAER will remove chloride ion according to this reaction: Cl + R OH R Cl + OH (1) where R represents the complex resin matrix. In Fig. 1(a), Part I refers to resins which have finished the above reaction (in the form of Cl); Part II refers to the resins which are working. This part is called an exchange zone. With the chloride solution continuously flowing through the column, the exchange zone moves down. When the capacity of the anion resin is exhausted (in the form of Cl), the chloride ions will completely break through the resins. Usually, one can monitor the conductivity of outwater of the anion-exchange tower to indirectly monitor the break through point, but when the content of sodium ion in the outflow water from prior cation-exchange tower of water treatment system increases unexpectedly (e.g. cation-exchange tower exhaustion),

2 832 ANALYTICAL SCIENCES MAY 2004, VOL. 20 sample water (inflow water) fiber Part I R-Cl type resin Part II Cl + R-OH R-Cl +OH R-Cl and R-OH type resins Part III R-OH type resin Exchange zone The concentration of Cl in the outflow water C0 Cx e a Equilibrium exchange capacity Breakthrough capacity b d time c breakthrough point outflow water ( a ) (b) Fig. 1 Schematic drawing of ion exchange process in an ion exchange resin column. (a) The ion exchange process in the column; (b) variation curve of concentration of chloride in outflow water from the column. C 0, concentration of Cl ion in the inlet water; C x, concentration of Cl ion in the outlet water; b point, breakthrough point; c point, the end point (completely exhausted point of the resin column). it will result in mistaken information of exhaustion of the anionexchange tower s exchange capicity. In fact, at this time, the exchange capicity of anion-exchange tower has not been exhausted. At power plants, chloride ions are the main cause of various metal corrosions. Its content also must be monitored and controlled strictly. Therefore, we selected the method of monitoring the chloride concentration in the outflow water. The chloride is determined according to the reactions below: Hg(SCN) 2 + 2Cl HgCl 2 + 2SCN (2) 3SCN + Fe 3+ Fe(SCN) 3 (red complex) (3) The absorbance of Fe(SCN) 3 is proportional to the concentration of chloride. So the concentration of chloride will be obtained as soon as the absorbance of Fe(SCN) 3 is measured at 465 nm. 11 Through detecting the chloride in the outflow water, we can obtain the outflow curve 12 that is shown in Fig. 1(b). At the beginning stage of the operation (from point a to point b), the concentration of chloride in the outflow water is low, because the chloride ions in the water are exchanged sufficiently with hydroxide ions on the resins when the sample water flows through the resins. At the point of b, chloride ions break through the resins in the minicolumn, so the curve rises sharply. Point b is called the breakthrough point. If the ion-exchange operation continues for some time, the resins in the column will be completely exhausted. Thus, the concentration of chloride ions detected in the outflow water will be equal to that detected in the inflow water. The value of breakthrough exchange capacity (BEC) of anionexchange resins should be calculated by the following equation: b E =Q/m = CdV (C0 C x)v= (C 0 C x)qt (4) m m m a where E refers to the breakthrough exchange capacity (mmol/g); m, Q, C, and V refer to the weight of resins in the column (g), the amount of ions absorbed by the exchange column (mmol), the differential concentration of chloride ions between the inflow water and the outflow water of the column (mmol/l), and the volume of water flowing through the column (L), respectively. C 0 and C x refer to the concentration of Cl in the inflow water and in the outflow water (mmol/l); t and q refer to operating time of the column (min) and volumetric flow rate of the sample water (L/min), respectively. Equation (4) shows that the breakthrough exchange capacity is proportional to the volume of the sample water flowing through the column as soon as the weight of resins remains constant. It will be used to calculate the exchange capacity in the later results and discussion. Apparatus A FIA-T1-721 Type flow-injection spectrophotometer with a flow cell (light path: 20 mm) made in the Instrumental Factory at Northeast China Institute of Electric Power Engineering 13 and a FIA-3100 Type flow-injection analyzer made in Wantuo Company (Beijing, China) were used. A XWT-100 Type recorder made in Dahua Meter Factory (Shanghai, China) was also used. Preparations of reagent and sample solutions All chemicals (bought from the Chemical Reagent Company of Beijing) used in the experiment were of analytical-reagent grade except that the sodium hydroxide, which was first-ratereagent grade. Ultra purified water whose conductivity is µs/cm was used for the preparation of these solutions. Chloride stock solution of mg/l. A g volume of sodium chloride heated for 2 h at 105 C was weighed, dissolved in water and diluted exactly to 1000 ml. Working standard solutions (20 mg/l Cl ) were prepared by appropriate dilution.

3 ANALYTICAL SCIENCES MAY 2004, VOL Fig. 2 Manifold for the determination of ppb-level chloride ions with FIA method. P, peristaltic pump; C, carrier stream; R, chromogenic reagent; RC, reaction coil; D, spectrophotometer; V, 3- channel synchronous/asynchronous injection valve; W, waste. Fig. 3 FIA manifold for the performance determination of ionexchange resins. P 1, P 2, peristaltic pump; C, carrier stream (purified water); IE, ion-exchange column; RC, reaction coil; D, spectrophotometer; V 1 and V 2, 2-channel synchronous injection valve; W, waste. Chromogenic reagent. The solution was prepared by dissolving g of mercuric thiocyanate and 30.3 g of ferric nitrate in water, with 3.4 ml of concentrated nitric acid and 150 ml of alcohol (95%) added later, and then diluted exactly to 1000 ml with water. Sodium hydroxide solution of 3.0% (w/v). A g volume of sodium hydroxide was weighed, dissolved in water, and then diluted exactly to 1000 ml. Preparation of ion-exchange mini column Preliminary treatment of resin x7 Type resins (equivalent to Amberlite IRA-400 Type resins (US) and Diaion SA-10A Type resins (JAP)) were weighed out to 0.15 g, immersed in the saturated sodium chloride solution for 12 h, and then immersed in 5% of hydrochloric acid and 3% of sodium hydroxide for 4 8 h, respectively. Finally, the resins were rinsed with ultrapurified water for 20 min. Filling of the ion-exchange column. The size of the ionexchange mini column was 3.0 mm inner diameter and 42 mm long. The treated resins were filled in the column, and two ends of the column were sealed with nylon net of 100 screen mesh to prevent the resins from flowing outside. In addition, some little fiber threads were put in the two ends of the column to prevent loosing or cracking of the resins caused by expansion and contraction. Then the column was connected in the flow manifold with a polytetrafluoroethylene tube (1.0 mm i.d.), as shown in Fig. 3. FIA manifold and analysis procedures Determination of chloride ion by FIA. First, rapid determination of ppb-level Cl was researched by FIA manifold, as shown in Fig. 2. Effects on the sensitivity of wavelength, temperature, length of reaction coils, pump speed, sampling volume, and concentration of the chromogenic reagent were investigated, and the optimum conditions of determining chloride ions were obtained. 15 Namely, the wavelength is 465 nm; the flow rate is 2.78 ml/min; the length of reaction coils is 300 cm (0.5 mm i.d.); and the sampling volume is 400 µl. The chromogenic reagent is the mixing solution of 0.063% (w/v) mercuric thiocyanate, 3.03% (w/v) iron(iii) nitrate, 15% (v/v) alcohol and 0.472% (w/v) nitric acid. The procedures are as follows: when the injection valve is switched to the loading position, the sample solution containing chloride is filled into the sample loop. Then the valve is switched to the injection position and the sample plug is injected into the carrier, and goes into the reaction coil (RC) after merging with the chromogenic reagent. The absorbance of the red product (Fe(SCN) 3) is measured at 465 nm while it is flowing through the flow cell. And the concentration of chloride is obtained by the peak height. The method s detection limit 16 is 20 µg/l, recoveries of adding standard sample are between 100% and 105%, the relative standard deviation is 0.89% for twelve determinations of 1.0 mg/l chloride solution, the sampling rate is up to samples/h. The method has been applied successfully to the determination of chloride ion in the boiled water at power plants, in natural water, tap water and so on. 15 In the study, by using the method, we have monitored chloride ion in the outflow water of the minicolumn to research the performance of the anion-exchange resin. Determination of exchange performance of anion-exchange resins by FIA. Figure 3 shows a FIA manifold used for the determination of ion-exchange resin performance. There is a 2- channel synchronous injection valve (including V 1 and V 2) in it. The procedures are as follows: first, under the optimum conditions mentioned above, a series of chloride standard solutions as the sample was determined, and the calibration curves were obtained as shown in Fig. 4. Second, the standard solution of 20 mg/l Cl is used as sample water. The sample water continuously flows through the minicolumn and enters into the loading loop of the valve (V 1). After the valve is switched to the injection position automatically, the sample water flows through V 2 to waste. At the same time, the sample plug in the loop is injected into the carrier and goes to the reaction coils (RC) and reacts with the reagent to form a complex (Fe(SCN) 3). The following procedures are the same as above. When the exchange capacity of the resins in the column has been exhausted (in the form of Cl), a running cycle of the column is finished. Consequently, a set of response peaks were obtained. Connecting these peak heights, we get an outline of outflow water of the column. Through this outline and calibration curves in Fig. 4, we can get information about chloride concentration, the breakthrough exchange capacity, equilibrium exchange capacity and resin exchange rate. During the whole experiment, the flow rate of the sample water, injection time, switch over of valve position etc. are controlled automatically by the computer. Besides, after the exchange capacity of the resins in the column has been exhausted, the column is regenerated with sodium hydroxide solution, and then rinsed with ultra-purified water for 10 min. Under the optimum conditions, according to the analytical procedures as mentioned above, we measured the exchange

4 834 ANALYTICAL SCIENCES MAY 2004, VOL. 20 Fig. 4 Calibration curves of chloride ions at higher and lower concentrations obtained with flow injection spectrophotometry. Experimental conditions are as follows: pump speed, 60 rpm; diameter of the pump tubing, φ0.89 mm; sampling volume, 400 µl; wave length, 480 nm; legnth of reaction coil, 300 cm; reaction temperature, 25 C; coloring reagent, 0.063% (w/v) mercuric thiocyanate % (w/v) iron(iii) nitrate + 15% (v/v) alcohol % (w/v) nitric acid. Fig. 5 Effluent curve of the anion-exchange resins (IER) determined by FIA spectrophotometry. Flow rate of the sample water, 1.5 ml/min; area of abde, breakthrough capacity; area of abcde, equilibrium exchange capacity; area of bcgf, the resin s exchange zone. Fig. 6 Effect of the regenerant dosage on the exchange capacity. performance of SBAER (the resins used were made in JY Manufactory). The results obtained are given in Fig. 5. It is found that the change tendency of the curve obtained by FIA corresponds with that of the theoretical curve in Fig. 1(b). In addition, such procedures were repeated twice under the same experimental conditions. The two curves are almost overlapping, which indicates that the method has high reproducibility. Results and Discussion Effect of regenerant dosage on breakthrough exchange capacity In the later experiments, exchange performance of anionexchange resins were investigated by the FIA manifold in Fig. 3. The regenerant dosage is an important factor affecting the regeneration efficiency of resin; thus, the effect of the volume of the regenerant on the breakthrough exchange capacity was investigated first. In the test, 2% (w/v) of sodium hydroxide solution was used as a regenerant, and the volumetric flow rates of both regenerant and sample water were 1.2 ml/min. The volume of the regenerant was controlled in the range of ml. The effect of volume of the regenerant on the resin s breakthrough exchange capacity was examined through changing the regeneration time. The experimental results are given in Fig. 6. They show that the exchange capacity increases with the increase of the regenerant volume, but the degree of increase begins to decrease when the volume is over 50 ml. Because it is not economical to increase the breakthrough exchange capacity by increasing the volume of regenerant, under the premise of ensuring the quality of outflow water to meet expectations, the volume of the regenerant should be chosen according to the economical principle. In this experiment, it was selected as 50 ml (0.15 g resin). Effect of regenerant concentration The volume of sodium hydroxide was fixed at 50 ml, and other experimental conditions were as mentioned above. The effect of the regenerant concentration on the breakthrough exchange capacity of resins was examined in the range from 0.5 to 10% (w/v). The results obtained are shown in Fig. 7. The exchange capacity increases initially, and then decreases as the concentration of the regenerant increases. There are two reasons to explain this phenomenon: first, the mole volume of OH is larger than that of Cl, so when the concentration of OH

5 ANALYTICAL SCIENCES MAY 2004, VOL Fig. 7 Effect of concentration of the regenerant on the exchange capacity. Fig. 9 Effect of volumetric flow rate of the sample water on the exchange capacity. Fig. 8 Effect of volumetric flow rate of the regenerant on the exchange capacity. Fig. 10 Effect of temperature of the exchange resins on the exchange capacity. is too high, the Cl will be absorbed prior to OH. Secondly, there is an electrical double layer at the resin solution interface. In the concentrated solution, the electrical double layer is contracted, so the resins are regenerated with difficulty. The greatest exchange capacity was achieved when the concentration of sodium hydroxide was 3%. Therefore, the concentration was selected as 3% (w/v) in this experiment. Effect of flow rate of regeneration solution The volumetric flow rate of regenerant exerts an influence on the regeneration efficiency, so its effect on the exchange capacity was also investigated in the range of ml/min. The experimental conditions were as in the section above. The results obtained are shown in Fig. 8. The breakthrough exchange capacity is inversely proportional to the volumetric flow rate of the regenerant, which means that the contacting time between the regenerant and resins is shortened when the flow rate increases. In this experiment, the volumetric flow rate was selected as 0.5 ml/min (4.3 m/h). Effect of flow rate of sample water The experimental conditions were as in the section above, except that the volumetric flow rate was 0.5 ml/min. Effect of volumetric flow rate of the sample water (20 mg/l chloride solution) on the resin capacity was examined in the range from 0.5 to 4.3 ml/min. The experimental results are shown in Fig. 9. The figure illustrates that the resin capacity remains nearly constant in the range of low flow rate of the sample water, and then decreases with the increase of the flow rate. When the flow rate is at 4.3 ml/min, the quality of outflow water becomes worse. Because the higher the rate of flow and the wider the resin exchange zone, the more chloride ions will break through the resins before exchanging with hydroxide ions once the width of the exchange zone is more than the length of the mini column. Based on the idea of both reducing the exchange time and increasing the exchange capacity, the volumetric flow rate of the sample water was selected as 1.5 ml/min (13 m/h). Effect of temperature Temperature is usually one of the most important influencing factors in chemical reactions, so in this experiment the effect of temperature on the resin performance was examined. The experimental conditions were as in the section above, except that the volumetric flow rate of the sample water was 1.5 ml/min. The mini column was put in the automatic thermostat bath of a FIA-T1 Type flow-injection analyzer. The

6 836 ANALYTICAL SCIENCES MAY 2004, VOL. 20 temperature of the resins was controlled by changing the water bath temperature in the range from 20 C to 80 C, while at the same time, the sample solution was heated to the same temperature as the column. The results obtained are shown as in Fig. 10. Results show that the exchange capacity increases as the temperature rises in the range of C, and decreases as the temperature rises in the range of C. It was found that when the temperature was at 80 C, the quality of outflow water became worse, the resin beads cracked and their color became deep. In the actual operation, the temperature should be kept in the range of C, because the physical degradation of resin will take place if the temperature is too high. Thus, in this experiment, the temperature was selected as room temperature, 25 ± 5 C. Conclusion Based on flow-injection spectrophotometric analytical technique, a method for determining the exchange performance of SBAER has been developed. Factors affecting the breakthrough exchange capacity of the resins were also investigated, and the optimum conditions were obtained. The wavelength is 465 nm; the flow rate of chromogenic reagent is 2.78 ml/min; and the length of the reaction coil is 300 cm (0.5 mm i.d.); the sampling volume is 400 µl; the chromogenic reagent consists of 0.063% (w/v) mercuric thiocyanate, 3.03% (w/v) iron(iii) nitrate, 15% (v/v) alcohol and 0.472% (w/v) nitric acid; the length of the ion-exchange minicolumn is 42 mm and the width is 3.0 mm. The amount of resins used is 0.15 g; the volume of the regenerant (sodium hydroxide) is 50 ml per 0.15 g resin, and its concentration is 3% (w/v). The volumetric flow rates of the regenerant and the sample water are 0.5 ml/min (4.3 m/h) and 1.5 ml/min (13 m/h), respectively. The exchanging temperature is 25 ± 5 C. This automatic method is characterized by the use of an ionexchange minicolumn, easy operation, high reproducibility, and rapid determination. Finishing the determination requires only 6 8 h by this method, in contrast to the duration of one week needed by the manual method. 4,5 So the method is approximately 30 times more efficient than the manual method, and it can be used to compare the exchange performance of various SBAER and to do theoretical research satisfactorily. In addition, some factors (such as the length and, the width of the column) affecting the breakthrough exchange capacity of resin were not examined in this experiment, so the value of breakthrough capacity is lower than that obtained by other methods. Thus, this method will be studied further. References 1. F. R. Jiang, Liaoning Chemical Industry, 1997, 26, Q. Zuo, S. F. Ran, J. Tang, and C. Y. Zhu, Ion Exchange and Adsorption, 1995, 11, B. S. Wang, Y. M. Liu, Y. G. Zhang, W. L. Li, and B. L. He, Ion Exchange and Adsorption, 1995, 11, ASTM D , Standard test method for operating performance of anion-exchange materials for strong acid removal, GB , Determination for exchange capacity of anion exchange resins, GB , Determination for exchange capacity of cation exchange resins, GB , Strong basic anion exchange resins in chloride form Determination of exchange capacity, Y.-S. Li and W.-C. Cheng, Flow Injection Analysis (Chinese), 1987, Beijing University Publishing House, No Y.-S. Li and Y. Narusawa, Bunseki Kagaku, 1992, 41, Y.-S. Li and Z.-J. Wang, Laboratory Robotics and Automation, 1995, 7, Y. Utsumi, Nikka, 1952, 73, X.-J. Shi, Treatment of water in power plants, 1993, Water Conservancy and Electric Power Publishing House, Y.-S. Li and W.-C. Cheng, Fenxi Yiqi, 1986, 3, J.-X. Yao, X.-W. Xu, and Y.-Y. Li, Industrial Boiler Water Treatment and Analysis of Water Quality, 1987, China Labor Publishing House, Y.-S. Li and Y.-L. Dong, North China Electric Power, 2003, 266, Z. Liu, Instrumental Analysis, 2001, Chemical Industry Publishing House, 228.